YT YouTube2Text

YouTube2Text → Transcripts → Supplemental lecture: Political Institutions and development

Supplemental lecture: Political Institutions and development — Transcript

by CEPR & VideoVox Economics · 11,957 words · 2,306 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:01welcome back to
0:02[Music]
0:04steggs virtual course on key concepts in
0:06macro development
0:08stegg is a cepr led research program
0:12structural transformation
0:14and economic growth uh
0:18with a focus on the macro development of
0:20low-income countries
0:23um it the c the
0:26steg program as well as this course are
0:29funded
0:31by the uk's development
0:34uh uk's aid effort i should say
0:38by the foreign commonwealth and
0:39development office so we're very
0:41thankful for them um
0:44today we have leonard wanchakun who will
0:47be
0:47talking about political
0:50economy political distortions and
0:53their relationship to structural uh
0:56their relation to economic development
0:59um we're thankful for leonard uh
1:02as i've mentioned many times these
1:04lecturers are doing this all out of the
1:06goodness of their heart
1:08let me tell you a little bit about the
1:09ground rules remind you the ground rules
1:11again
1:12while people are flooding in um
1:16the front row you can
1:19unmute if you have a clarifying question
1:21during the talk
1:23uh leonard will give some time at the
1:24end for you know broader questions
1:27but if there's some clarifying question
1:28please unmute uh otherwise leave your
1:31microphones muted
1:33those who are not panelists who are just
1:36participants in the webinar
1:38can ask questions in the q a i will do
1:41my best to try to answer what i can
1:43although this is not my area of
1:45expertise so i'm a little bit less
1:46confident than i've been in previous
1:48lectures
1:49and doug is not here to help me if we
1:51can't get
1:52questions answered we may try to leave
1:54those for leonard
1:55uh during the q a time
1:58[Music]
2:00and the lectures the slides themselves
2:04are available at the course website
2:07and with that um let me uh
2:10hand things over to leonard yeah thanks
2:14joe
2:14thanks uh cpr for for the opportunity
2:19um you know i'm extremely um
2:22actually very grateful that i'm i'm
2:24given the platform
2:27to talk about this topic um
2:31so before before i start
2:34let me just tell you that studying
2:37political mechanism
2:39in economic development was my original
2:40interest so
2:42um so very very glad to have to come
2:45back to
2:47what i have always been interested in
2:49doing so
2:50um i'm going to start
2:54sharing my screen um
2:57immediately so the talk uh will be
3:00structured this way first of all i'm
3:03going to
3:04introduce the lecture
3:07by referring mostly to past lectures and
3:11how
3:13our discussion here will
3:16link to what had been covered earlier
3:20and then i'll move to some theoretical
3:23background which is important because
3:27we need to sort of set some
3:30benchmark uh benchmark model
3:34to really understand the role of
3:37political institutions
3:39and political distortion on economic
3:41development
3:42now later on i'm going to cover
3:47like three type of follicle distortion
3:51uh departure from the
3:55benchmark model that we are going to be
3:57covering
3:59one has to do with distortion of public
4:02investment
4:04that's usually called patronage
4:08distortion of human resources
4:11in governments other type of
4:15targeted retribution which is distortion
4:18of
4:18all the type of public goods that the
4:20government might provide
4:23so and then following that we're going
4:25to talk about contract enforcement which
4:27is
4:28maybe the core components of political
4:31institutions because it has to do with
4:33property rights to have to do with law
4:34enforcement and so on
4:36um and then the third type of political
4:40distortion would be
4:41state capture you know like a
4:43relationship between firms
4:45and politicians so
4:48um to basically uh
4:52explain the mechanism through which
4:55those distortions take place especially
4:58the last one which is uh
5:00state capture uh basically
5:03we are going to have
5:06discuss we have to talk about lobbying
5:08you know for instance how
5:10lobbying by firm might lead to
5:12[Music]
5:14like create connections that might
5:20create political distortion of the third
5:21type
5:23later on i'm going to basically
5:28provide a remark on elite capture or
5:32create capitalism
5:33which is basically perceived
5:36as structural political destruction like
5:39when
5:40um political distortion is very much
5:42embedded
5:43it's very much into the structure of the
5:45way
5:46of the economy work and and uh
5:49and then talk about future directions i
5:50mean the the reason why this um
5:53i'm going in the lecture is that because
5:55there is so much left to be done
5:57you know there is a very important
6:00interesting topic but it's a bits all
6:03over the places so
6:04i think it's important that we
6:06understand
6:09how this literature can be put can be
6:12pushed forward
6:13okay so that's basically the wrong mark
6:16now as your introduction so
6:18let me remind you you know what you
6:19already know which is that um
6:22difference in living standard across
6:23countries are linked to growth
6:26driven by productivity gap and
6:29misallocation
6:31and that total factor productivity
6:35account for much of these disparities
6:38much of this gap
6:40you know i mean disparity in capital
6:41labor i come from mostly at most 50
6:44of differences in income per capita
6:47so okay so far the focus has been on
6:50market sources of differential and
6:52productivity
6:53you know like for instance mark a great
6:56lecture
6:56like two maybe a week ago or two weeks
6:59ago
7:00provide detail on labor market frictions
7:03and the effect on development
7:04he talked about frictions in
7:06agricultural market
7:08authentic firms differences in hiring
7:10costs across farm and technology
7:12adoption
7:14and but it is in this section so
7:17what we are interested in is not the
7:19market mechanism but the political
7:21mechanism
7:22for political distortions and
7:25particularly
7:26the effects on human capital formation
7:28and innovation
7:30on capital investment and on
7:34special friction for instance and
7:37misallocation
7:40and you know you know
7:43last lecture by uh roger i mean by
7:46richard i'm sorry
7:47rogerson um and actually the first
7:50lecture he
7:51uh explicitly incorporated uh um you
7:55know
7:56differences in one sector group model
8:00and a heart it
8:03is the relative position of a country in
8:06technological from the technological
8:08frontier
8:09and mentioned in passing that this
8:12can be affected by country-specific
8:15policies and institutions
8:17and you know i mean like foreign
8:19institutions might affect
8:21the level of capital through
8:25infrastructure investment um
8:28you know you might make more
8:29international investment you might may
8:31you might invest more in capital for
8:33instance
8:34if um you know you the government
8:37provided
8:38energy or provided roads you know that
8:40makes uh that improve
8:42the productivity of your capital
8:43investment and it could also be
8:46through subsidizing public education for
8:48instance you know
8:49that will affect l so one of the
8:53one of the key i mean one of the key
8:55areas that
8:56the roger in his paper with restitution
9:00rogerson and people restructure covered
9:03is uh you know how the allocation of
9:06resources across
9:07for instance iteration plans how
9:09important they are
9:10to explain cross-country differences
9:14and the fact that taxes and subsidies
9:18for instance
9:20can create uh in fact misallocation
9:26and um and now now
9:29the the the natural you know the natural
9:32question
9:33uh from that that setup
9:36would be where those where do where
9:39where does thai tau i come from so it
9:43could be correlated for instance with
9:44firms
9:45political connections you know like for
9:46instance some firms
9:48benefited benefit from higher from lower
9:51tax rate than others
9:53because they basically either pay bribe
9:55to politicians or
9:57they they make campaign contribution to
9:59the politicians so
10:01so political connection for instance as
10:03we'll see later on
10:04my my drive means allocation and and and
10:08and growth rate in a very very
10:10fundamental way
10:12okay so now since this paper i mean this
10:15lecture is about political institutions
10:17so uh what are the what happened what do
10:20you what do we mean by political
10:21institutions
10:22or institutions in general so um
10:26it's probably one of the most
10:30one of the broadest concepts in all
10:32social science
10:33uh because the terms uh
10:36institutions are basically you know
10:40all over the places you know so uh but
10:42then
10:43let's focus on four that i think i'm
10:48quite you know uh insightful
10:51so you have dognoff a definition which
10:54uh focus on you know
10:58rules of the game you know
11:01which is like constraints that shape
11:03human interactions
11:06so myosin you know jameson
11:09consider those rules more like an
11:12incentive systems
11:14that shape political decisions and
11:17decides
11:19which strategies are followed
11:22or not you know i assimilate focus on uh
11:26robinson johnson robinson focused on
11:28property rights and contract enforcement
11:30now um
11:32[Music]
11:34eczewowski you know a political
11:36economist from nyu
11:38uh focus more on
11:42what political actors
11:46or social actors are not
11:50allowed to do you know like the checking
11:52balances
11:53that prevents um
11:56you know basically uh uh key
12:00political actors to um to
12:04you know to to violent
12:07that not to violate norms of
12:11of ecosystem particularly democratic
12:13system
12:14so now for the purpose of this lecture
12:18i will just in a nutshell
12:22consider institutions at
12:25three things they are mechanism
12:29for making policy decisions basically
12:33it's a technology it's governments
12:36that make policy decisions that's that's
12:39what it is
12:41so it's an organization a government
12:43organization
12:44the second is institutions
12:47is about contract enforcement mechanism
12:50okay and the third is a mechanism for
12:53selecting
12:54public office holders so as a result
12:57institutions is institutional
13:01make rules you know decides
13:04you know public policy and false
13:08rules and uh a way to decide
13:11who make rules you know so and those are
13:15really different um uh set of
13:18um mechanism so not surprisingly
13:23the model that capture the effect of
13:25institutions on economic development
13:28will vary depending on whether
13:31you are talking about the first the
13:33second
13:34and the third mechanism you know so
13:38it's always important when you want to
13:41link
13:41up institutions to development
13:44that you refer to the make
13:48to one or the other mechanism described
13:51here
13:51okay so so that's exactly what we are
13:55going to do we are going to
13:57to to sort of uh go over
14:00some benchmark growth model that
14:03capture specific aspects of what we mean
14:06by
14:07institutions so for instance the first
14:10one
14:11government government policy
14:14public investment that affect growth
14:18so that's barrow 1990
14:21and also barrow and sari martin
14:261994 or maybe later i forgot the dates
14:30and then the second one the control
14:32enforcement we are going to cover
14:34agia and amador's paper
14:37where you know um you have weak
14:41basically enforcement of the rules
14:44and then in terms of how
14:48the the way political selection
14:52affect particularly innovation
14:57and productivity so we are going to
15:00cover the strong paterian
15:01model that aguero and al and asimo blue
15:04and now has
15:06developed in several papers so and then
15:08so this is the benchmark
15:10you know so um and the most the most
15:13crude and the most basic model of
15:16government and political institutions
15:18will be the first one you know the
15:19barrels model and then the other one
15:21will be some extensions or
15:25way to expand and to deepen
15:28uh you know barrow 1990 and 94.
15:32so and having done that then what we are
15:36going to do
15:36is to present departure from this man
15:38landmark you know
15:40any any um
15:42[Music]
15:43institutional effects or institutional
15:46features
15:47that depart from those benchmark model
15:49it's what we are going to call
15:50political political distortion so and
15:53this is really really
15:54important because you know
15:57when we want to evaluate for instance
16:00institutions
16:01it has to be in relationship
16:04with a benchmark you know we have an
16:08idea
16:09of an optimal government
16:12or optimal government allocation of
16:14actions or policies
16:17that you know either because you know
16:20the government in maximizing
16:22uh social welfare or maximizing um the
16:25utility function of representative
16:27agents that represent the interest of
16:30the country
16:32and anything that departs
16:35from the prediction of this basic model
16:39will be considered political distortion
16:41you know would be
16:42and i think it's interesting because it
16:44helped
16:46get some parallel between the way we
16:48look at institutions
16:51um you know about political institutions
16:54um between that and the way we look at
16:56economic institutions
16:58you know anyway so so i think it's
17:00that's very very important
17:01okay now and then later on we are going
17:04to document how institutions
17:07lead to um you know political distortion
17:10um you know by looking at particularly
17:13public investment and patronage
17:15and looking at political connection and
17:16state capture and
17:18lobbying and misallocation so there are
17:20a large number of papers
17:22i'm pretty sure that we don't have time
17:23to cover all of them that's why
17:25you know i i mean and purposely
17:28you know i made the the slide
17:32very very long so that you can actually
17:34look at those paper yourself
17:35or come to the office hours next week so
17:38that we can go over them with more
17:39details okay
17:41so so let me then briefly uh cover
17:45the benchmark model so basically it's a
17:47under your model that include public
17:49services
17:50you know as a productive input and
17:53that's productive input
17:55um can be a publicly provided public
17:59goods
17:59that are rival or equitable it can be
18:02also
18:03publicly provided public good which are
18:04non-rival non-excludable
18:06it could also be um publicly provided
18:09goods
18:10that are that are consent that that's
18:12subject to congestion
18:14you know um so that highway
18:17water sewage is for instance so so
18:20basically
18:21uh here the government is more like a
18:23technology
18:24it's basically like a technology and um
18:30and um and as i said earlier on
18:33you know the government's role is to to
18:35maximize the utility functionality of
18:37households and there are
18:40a number of uh producers
18:44there is you have the capital and you
18:46have government spending
18:48as key input and they aggregate to
18:51capital g
18:53and and now the key is that government
18:55finance is spending
18:56their taxes so and then they look at uh
19:00several scenarios you know one
19:04is for instance this one you know where
19:07as you see the government spending
19:09coming
19:10as a private private like maybe
19:13a private private input publicly
19:15provided private input
19:17and and then you see that uh you know
19:21financed by by
19:24by taxes and and then
19:27you know derive
19:30equilibrium and show that in in this
19:34model
19:35only unless tax is long some
19:40it's it's distortionary you know
19:43okay so the the the government's
19:47action create distortion in the economy
19:51um because unless
19:54you know unless uh the tax rate is lump
19:57sum so that's one result
19:59the second one is the publicly provided
20:01public goods
20:02it's in that case you have
20:05capital g which enter the utility
20:07function of each
20:09producer and in that case the results
20:12are i mean the result is the same as
20:14before
20:15but when you have conjecture you know
20:18when for instance
20:19why um you know
20:22government uh government expenditure
20:25enter
20:25like this the conjunction the congestion
20:29variable is measured by this then
20:32the conclusion is that uh tax
20:37are not are not necessarily this
20:40decisionary
20:41and they are desirable you know so so
20:44that's basically
20:45the the minute but what we need to
20:47really retain
20:48what we need to remember from this
20:50exercise
20:52is the role of governments which is a
20:53technology
20:55and then um the fact that um
20:59um you know the the the the
21:02the fact that tax creates a wedge
21:06and if this it's decisionary unless for
21:08instance
21:09um you know what's being provided is um
21:13you know the public good provided is uh
21:17his projection okay so that's uh
21:20that's roughly about um uh
21:24barrow's uh setup now assume for
21:27instance
21:28that um you don't have the government's
21:33is not one
21:36kind i mean the government is uh has
21:40is divided you know that you have um
21:43someone in power someone outside of
21:45power waiting to get into
21:47um into power i mean into into
21:50government
21:50and that those in power cannot commit
21:54uh they cannot commit not to appropriate
21:56you know private producers
21:58and they also cannot commit not to
22:01default
22:02on on their death and it has some
22:05it is also assumed that they overvalue
22:07present consumptions
22:09being being power for instance it's
22:11something
22:12which is uh you know which uh
22:15give whomever in power some kind of uh
22:18present or some kind of
22:22bias towards current
22:25and in that setup with limited
22:28commitments
22:30and they show that political
22:33disagreement
22:34which is the gap in the in
22:37the gap in payoff between
22:41those in power and those who are out of
22:43power you know the
22:44level of polarization between
22:48um those who are currently in power and
22:51those who are not
22:52the more disagreement with the bigger is
22:56the
22:57degree of polarization between between
23:00them
23:00measured by a parameter
23:06data the
23:09growth will be slower you know so
23:11basically in this building setup
23:13they show how those in power have
23:17trouble committing to do the right thing
23:20and that ones that not doing the
23:24being power is highly rewarded rewarded
23:28compared to not being power then
23:31what's going to happen then is growth
23:33the speed of convergence instead it is
23:37i mean the set of convergence will be
23:39will be
23:40would be slower okay so that's basically
23:43how
23:45um you know how um like
23:48uh you know political competition
23:52and polarization for instance can play a
23:55role
23:56in slowing down growth rate
24:00especially in environments where you
24:02know
24:03those in power cannot commit not to
24:06appropriate
24:07and not to increase death okay
24:11now one of the most influential
24:14um political economy uh studies that
24:17link
24:18um basically political institutions to
24:21to
24:21to development with the schumpeterian
24:23criterion model
24:25that asks a very basic question which is
24:28you know
24:28is does democracy enhance or hamper
24:32growth there are theories showing it
24:35does because
24:36there is greater greater pressure for
24:38redistribution
24:40and you want to reduce a bit to get
24:41elected as a result this
24:43create incentive this disincentive for
24:46capital formation
24:48other talk about democracy
24:51can increase you know growth
24:54because it protects property rights
24:58and and lower the ability
25:01of incumbent to block new entrants and
25:04as a result
25:05uh to block innovation so that those are
25:08the two
25:08trends and and uh schumpeter's superior
25:11model focused more on the
25:13on the ladder you know on um the
25:17you know basically how democracy can
25:21can help and can promote uh innovation
25:24and growth
25:26and but then um in the model innovation
25:29and growth
25:30are promoted in countries
25:34closer to the technology frontier so
25:38countries at the technology configure
25:41like very very developed
25:42countries for instance in those
25:43countries then
25:45um democracy will be
25:49um will promote innovation
25:52and growth um and
25:56the the the model is very very stylized
25:59you know
25:59so we assume that you know output depend
26:02on technology
26:04and there are advanced sectors
26:07there are also backward sectors where
26:10you know the technology from yesterday
26:11is the
26:12technology today and then in each sector
26:15you have one incumbent and one potential
26:18entrance
26:20and and then we assume that only the
26:22potential
26:23entrant innovates
26:26but then before production potential
26:29entrant has to invest in ind
26:32in order to replace the incumbent and
26:35if successful productivity increased by
26:38this factor
26:40and then this will make the new entrance
26:41monopolist
26:43otherwise the current incumbent remain
26:46and
26:47you know things remain kind of
26:50backward if you will okay so
26:53now the the way democracy is introduced
26:56is simply through the probability
27:00that a successful innovation will lead
27:02to successful entry
27:04you know democracy is just the
27:06probability
27:08that um a new entrant
27:11will be able i mean a new firm
27:15innovative firm would be able to enter
27:18um you know uh the market so it's a it's
27:20a
27:22the assumption that's made here is that
27:23there is very very tight connection
27:26between the governments and um
27:29and the incumbent firms between
27:32incumbents in government
27:33and incumbent firms so and
27:36uh and as a result uh the the fortune
27:41of the new entrant is tied to the
27:44probability
27:45that i depend very heavily on the
27:48probability
27:49that whomever is in power can be
27:51replaced
27:52so so the key aspect of democracy here
27:55is the ability that it gives to an
27:58entrant
27:58politicians or firms to enter the market
28:02you know so if the potential
28:05entrant um you know as i said uh
28:09earlier they spend this
28:12much they can innovate with this
28:16probability
28:17so as a result the joint probability of
28:19an unblock entry is given by this
28:22so very stylized this is the the
28:25maximization problem
28:26the benefit and the cost and this is
28:28monopoly profit
28:30and in equilibrium this is what's
28:32happened and then you can see
28:34that you know the innovation effort is
28:36increasing
28:37in level of democracy and profit
28:40decreasing
28:40in rd cost okay so
28:44now if you look at the average uh level
28:46of productivity in the economy
28:48based on this it looked like
28:51this so which means that
28:55you know this term the aggregate product
28:58the average productivity growth
29:00in the economy is increasing level of
29:02democracy
29:03and democracy is more enhancing
29:07uh in in in frontier
29:10uh you know
29:13when uh when the country is close to the
29:16frontier and
29:17and uh aguila alessina and trevi has uh
29:20provided
29:21uh cross-country empirical evidence on
29:23this
29:24um that you know we can cover later on
29:27okay
29:27so this is um so so that's about
29:31one version of uh the schumacherian
29:35growth if you will but then another
29:38version or another extension
29:40of the original model is by asia can i
29:43ask something
29:44yes go ahead please uh maybe i missed uh
29:46missed it but uh in
29:48in the assumptions we made does it not
29:50mean that
29:51every time there are like uh elections
29:54the government will change
29:55because innovation comes only if uh the
29:58government changes
30:00or is there something which which will
30:01ensure that is there a cost as well
30:05so so with okay um
30:09maybe ask your question again i'm i'm
30:11not sure if i how you understand it
30:13properly go ahead
30:14uh in the assumptions that we made uh
30:16before this
30:17yeah i was just wondering that uh
30:19because only if the government changes
30:21is their
30:22uh innovation so is there anything which
30:25ensures that
30:27something like a cost which ensures that
30:29there's a possibility that the
30:30government does not change
30:32that's a very that's a very very good
30:34question so first of all um
30:36here it's not necessary okay let me put
30:38this way
30:39is the possibility that the government
30:42will change
30:43that help discipline
30:46the incumbent you know so in actuality
30:50it might be the case that the government
30:51stay
30:52but the fact that they
30:56they um you know that they
31:00that they are threatened exit
31:03to to exit my discipline then
31:06not to block technological change
31:10but empirically the evidence
31:13show is that when
31:16incumbent does lose election
31:19you know when the possibility exists
31:22that incumbent does lose
31:24power and election when political
31:27competition
31:28is is real then you have more
31:32innovation at the top at the frontier
31:34you see what i mean
31:35so the point here you can make the case
31:37that actual turnover in government
31:39help produce productivity and help
31:42innovation
31:44but but even
31:47but perhaps i don't know this was shown
31:49explicitly
31:50in in in in this is shown empirically
31:55but the threats to lose election
31:58might also create um
32:01incentives for innovation
32:06you know because the incumbent has some
32:08protection
32:09okay so okay uh sir excuse me
32:13yeah okay uh leona um is that not
32:16also important that we have
32:20the right definition of democracy in
32:22this regard because the assumption for
32:24me it's uh
32:25it's too broad some might argue that
32:29political distortion i mean
32:32instability in government or possibility
32:35of government changing could also in the
32:37investment
32:38in that regard so i think i would want a
32:41what
32:41in what regard i mean what is the
32:43population of democracy in this case is
32:44it yeah yeah
32:45okay democracy
32:48i understand the question so okay let's
32:50distinguish between
32:52uh states and government okay okay
32:56okay if states are strong
33:00and stable meaning there are rules
33:05governing the way investment should be
33:08made
33:09public goods should be distributed and
33:11those rules are applied
33:13regardless who is in power there is no
33:16reason why
33:17they should deter investment okay okay
33:20that states state stability and
33:24state capacity okay but then you know
33:28for instance
33:28i agree with you that whenever you have
33:31a situation where
33:33governments turn over weaken
33:36states in if
33:40you know from the city government
33:42turnover for instance can deter
33:44investment
33:45you know so for instance there are many
33:46stories showing for instance that a
33:49local level
33:51when mayors lose election
33:54um it has bad impact it has low
33:57it has a negative impact on learning
34:00outcome at school
34:02because when a mayor leaves the new
34:05the new one come fire everyone else
34:08come with a new staff
34:12and it's back to square one you see what
34:14i mean so
34:15so so i think the point you are making
34:17is a very valid one we have to make sure
34:20that the the stability that the states
34:25remain stable in spite of its bias
34:29of government turnover exactly this map
34:34that's a very very important point so
34:36yeah anyway
34:37thank you um and so so yeah and then um
34:41i see morgan robinson for instance i
34:43mentioned earlier they looked at
34:45how incumbent can block technological
34:48change
34:49uh because uh innovation will reduce
34:53uh their power you know basically they
34:55are too channeled here
34:56one channel is that one channel where
34:59this happens
35:00is that the firms and the government
35:03are so tight that when
35:07the firm when the government leaves
35:11the firms is weakened and
35:15the resources that the government get
35:18disappears you see what i mean so as a
35:21result they make sure
35:23that even that that connected firms
35:27remain you see what i mean so that's one
35:29the second
35:30is that technological change for
35:32instance can empower citizens
35:35you know it can make you know citizens
35:37for instance
35:38uh more powerful and powerful citizens
35:42will demand
35:43change you know you see what i mean like
35:45uh you know i mean a government that
35:47enable
35:48digital or social media for instance
35:52um might uh be working
35:55against its interest you know because
35:58citizen would be better informed and as
35:59a result they are going to demand change
36:01okay so so clearly the idea here is that
36:04technology might be blocked because
36:07governments in power might
36:09fear how these are going to affect the
36:11decision that citizens will make
36:13and it also the fact that they might be
36:16so tightly connected to um to
36:19incumbent firms that and they don't want
36:22this firm to to to
36:23to to leave okay so so that's uh
36:27that's the asimoglou and robinson
36:29johnson robinson
36:30version of how
36:33um democracy will promote growth
36:37by blocked by
36:40by punishing or by
36:43incumbents that have an incentive to
36:46block
36:46uh technology because of their economic
36:49ties okay
36:50so okay this is those are the benchmarks
36:53you know one
36:54is about government turnover the one is
36:55about control enforcement the
36:57first one is uh public investment those
36:59are the three
37:00set of um you know three setup um
37:04institutions okay so now in this videos
37:08very simple environment
37:10so let's look at um how
37:13political institutions in this simple
37:15environment
37:16will affect growth and consumption
37:20debt profile you know coming from uh
37:22aggia and
37:23um uh and amador technologically in in
37:26innovation firm productivity and
37:30and dynamics uh coming from asimoglou
37:33and now arguing now okay so that's
37:36basically
37:36what we are going to do as i mentioned
37:38earlier we just take this
37:40as a benchmark and say okay how does the
37:43deviation from the prediction of this
37:46benchmark model for instance
37:48would be detrimental to growth so
37:51what you know we want to see whether
37:53this effect can be observed from
37:54critically
37:56um you know which firm will benefit and
37:59lose
38:00um and how to think about the choice of
38:03political connections
38:04and how the type of policy
38:08that we should consider to improve
38:09welfare and so on
38:11so so basically we are now
38:14basically going michael you know to see
38:19how um
38:22um connections of different type
38:25okay because remember you know in
38:28barrow politicians just wanted to i mean
38:31the government just wanted to
38:33maximize the representative
38:36consumers equity function but government
38:39has to get elected
38:41you know and what they want instead is
38:44how to maximize their probability of
38:45being elected
38:47how we want to uh maximize the payoffs
38:50coming from being elected you know not
38:53uh
38:54you know anyway so so if we look at this
38:58and then uh then we can understand
39:01political connection and we can see how
39:02those connections for instance
39:04will will distort the optimal allocation
39:08of resources that government will
39:10provide okay
39:11so we are going to take several examples
39:14the first example is patronage and
39:16target reciprocal
39:17distribution which is mostly like
39:19distortion of public investment
39:21so we are going i'm going to go through
39:22two examples and maybe many more but you
39:25know
39:25feel free to actually read more about it
39:28because i think it's
39:28very very interesting the first one is
39:31um
39:33basically the allocation of roads
39:37in kenya
39:40and it shows that
39:44democracy redistributes
39:48more almost more or less fairly
39:52um between a road network between
39:57areas that are dominated by co-ethnics
40:00and area dominated by non-coherent
40:03you know so and
40:07so if we want to choose between
40:10democracy
40:11and dictatorship between
40:14turnover in governments and not
40:18the claim which is made here is that
40:22democracy is better not not
40:25not that democracy is perfect
40:28you know but it's better than autocracy
40:32you know and um and they look at
40:35basically the effect of road expenditure
40:38on co-ethnic
40:39districts and the effects
40:43and the show that the effect
40:45misallocation
40:47or distortion in public investment
40:50giving more i'm going to be explaining
40:53about distortion what what distortion
40:55mean in a second
40:56giving more to queue ethnics and
40:58co-ethnics it's likely to be present
41:01on the autocracy but disappear on the
41:03democracy
41:05okay so there are similar stories
41:09done um in this time
41:13not between autocracy and demo and
41:15dictatorship but within
41:17democracy and show for instance how
41:20once office holders want to hold on to
41:23power
41:25even in a democracy they can misallocate
41:29public investment you know they can
41:32you know in india they will provide
41:34insurance to vulnerable clients for
41:36political favor
41:37in pakistan in uh sabrina's work
41:41berg's work they show how for instance
41:44um they um you know
41:48how politics affect um
41:52tenancy and and and there are a lot of
41:56example also in the us okay
41:58so here is here is basically um the
42:01picture here
42:02so interestingly what is interesting
42:05here
42:05is that you have a
42:08the you you usually evolution of um
42:13of you know you you switch
42:16in kenya between democracy and
42:18dictatorship
42:19or directorship and
42:23back to democracy and look at
42:27what whomever's in power do when
42:32uh does i'm sorry when the
42:35when you have political transition
42:38so for instance here
42:41um it was a kikuyu dominated
42:45period i mean it was it was uh you know
42:47it was
42:48a democratic democratic government
42:51autocracy um and then
42:55autocracy back to autocracy back to
42:58democracy
42:59and you know and and i'm not sure
43:02between
43:0310 and up but you know it's uh i think
43:06it's back to democracy
43:07anyway but the point here is that as you
43:10can see
43:11in period of democracy like here
43:15and here you have oh yeah particularly
43:19particularly here
43:20you have a higher
43:24co-ethnic bias sorry you have a lower
43:27ethnic bias in road allocation than on
43:30the autocracy
43:31you can see here for instance a location
43:34of public investment in roads
43:36much higher you know um
43:40under on the autocracies done on the
43:43on the on on the democracy okay so
43:47and anyway so and and for instance
43:50[Music]
43:52this is confirmed by this table and
43:55to show that this is dictionary they
43:57show
43:58they look at the contrafactual world
44:00network
44:02the world network that a a social
44:05planner
44:06or the a
44:10development outcome maximizing
44:13government will choose
44:14and compare it to the actual road
44:16network and
44:18and show that uh under democracy
44:21there is less gap in terms of uh
44:24the less i mean smaller gap in between
44:28the actual road network and the
44:29contractor contractual road network
44:31than there is on the autocracy okay so
44:34that's one
44:35about destruction of public investment
44:38in roads
44:38you know there are a lot of studies
44:40showing similar distortion in
44:41educational investments
44:43and um and also in
44:47you know income distribution scheme that
44:49might be that might take place
44:50okay and again that i said this was a
44:53comparison between democracy and
44:54dictatorship
44:55but in fact if you look within
44:57democracies for instance
44:59you can see that
45:03this you know um politicians
45:07might favor
45:11some specific groups more likely to vote
45:14for them
45:16you know than other groups anyway so but
45:18this is just to make a point
45:20that institutions
45:23democracy and also within democracy
45:27um you know and the the
45:31can lead to that can lead to distortion
45:34in in public investment
45:36now the second part the second type of
45:39distortion
45:40is patronage so what is spectrometer is
45:42the social reappointment of individual
45:44to government and political position
45:47basically is
45:51filling government position with
45:56political appointees you know okay and
45:59this
46:00can happen two ways it can happen um
46:03as if type of political exchange
46:05exchange
46:06you know you helped me get elected
46:10and then i i asked you know i put in
46:14charge
46:15of a government agency
46:18it could also be um
46:21driven by social ties you know
46:25in a sense like um you are
46:28from we went to school together
46:32you are one of my you know you are in my
46:35social network
46:36and i reward you by putting you
46:39in a government position you know so so
46:42again the ties here can be political it
46:44can be social
46:45all right so close connection of firms
46:49and states in hiring practices
46:51my effect who get who get fired who
46:54could get hired
46:55can affect the quality of bureaucracy
46:57and policy implementation
46:58and which policy is implemented so there
47:00is a classical
47:01classic studies done uh by sue
47:04in 2018 about for british empire
47:08and basically the context is british
47:10empire 1854 and 1966
47:14with they used a newly digitized data
47:18and um we study we study the effect of
47:21patronage
47:22that i defined earlier on promotion to
47:25higher salary position
47:27and the
47:30and the effect on tax revenue tax agents
47:33and exemptions
47:34and an investment okay and they they
47:37study use two type of variation
47:39the one is electroshock because
47:43first of all a party has to win
47:48to be able at the party you have to win
47:51to select
47:52the secretary of state i mean foreign
47:55minister
47:56who then will select a governor
48:00you know so so uncertainty about being
48:04elected in that position is
48:05the first source of variation the second
48:08source of variation
48:09is is the fact that in 1930
48:13there is the warren official reform that
48:15makes this appointment overseen
48:17by independent civil service civil
48:19service
48:20so so i mean the reform makes
48:23patron appointment less likely okay
48:27and this picture tell the story very
48:29clearly you can see
48:30for instance that in terms of average
48:34salary
48:36political appointees was making more
48:39than non-political appointees so
48:42patronage
48:44increased the wage bill for the
48:46government but then after 1930
48:50you know from 1930 the gap
48:53um decreased significantly to the point
48:57where
48:57it's more or less the same uh you know
48:59from 1950
49:00on until the british empire disappeared
49:04okay so basically um patronage is a
49:07distortion
49:08of public management you know and
49:12they also show for instance that um you
49:16know
49:16this has a negative effect
49:20on um on public expenditure on
49:22investment as i said earlier
49:25and um you know anyway so
49:28i will really strongly recommend to you
49:30to read the paper because
49:32uh you know the evidence is extremely
49:34extremely
49:35extremely extremely solid okay so now
49:38this is
49:39about um patronage you know how
49:42this institution might distort
49:46public investment in roads and distorts
49:49as well um public management
49:53you know through patronage and how this
49:55can lower
49:56tax revenues and
50:00and increase the cost of running
50:03governments you know through the highest
50:06salary for the appointees
50:07and so on you know there are even much
50:10worse form
50:11of patronage than this and in political
50:13science they call it
50:14pre-vandalism which is basically the
50:17political appointee
50:19not only gain by being appointed
50:23in terms of salary even if it is not
50:27as competent but sometimes
50:31they are given ownership
50:34of part of government and they can
50:37basically
50:38run it as they wish you know run it down
50:41completely
50:42okay so this is the worst
50:46so what i'm saying is that the evidence
50:47will be hanky we find him in terms of uh
50:50um in terms of um the cost of patronage
50:53described
50:54by tsu in in the british empire can be
50:57actually much much worse than that
50:58okay so okay now let's talk about
51:01control enforcement control enforcement
51:03i'm going to cover a question
51:06yes go ahead um before you go to the
51:09next one
51:09um quickly i think i it's about
51:12um an earlier slide which which is about
51:15distortion resulting from
51:17um the provision of public goods yeah
51:20yeah
51:20you mean yes yes yes yeah yeah yeah this
51:24one
51:25is is it is it the case that governments
51:28increase spending during elections could
51:30also be a source of
51:31this type of distortion so that
51:34government spends more during an
51:36election because it wants to be retained
51:39could that also be a source of this type
51:41of distortion absolutely
51:42absolutely and there are a lot of
51:43studies on that for instance um
51:46uh you know there are a lot of studies
51:48on for instance
51:50it's called a political business cycle
51:52you know
51:53where government for instance
51:57um over over spain
52:00at the eve of election basically
52:04in strategic areas you know so that they
52:07can retain
52:08uh power you know and it could be
52:11not only public investments you know
52:14like public investment in roads and
52:16infrastructure and so on
52:17it could also be patronage like for
52:19instance
52:20assume that you know you need votes
52:24from a given district in northeast
52:27to win the election you know that you
52:30are in power
52:31what do you do you appoint somebody from
52:34that region
52:35before the election to an important
52:36position
52:38by telling everyone well you know what
52:41if i lose the election
52:42your guy is going to be fired because
52:44i'm not there anymore
52:46but if i win yeah he's there so as a
52:49result you
52:50tie them in and force them to vote for
52:52you you see
52:53so so so you know i mean like those are
52:55just examples
52:57of distortion but i think i mean okay
53:00for those interested in political
53:02economy and growth
53:03this is some of the hottest areas of uh
53:06for research
53:07because the data is there you know and
53:10also because
53:10it's very powerful evidence
53:14of political distortion
53:18you know and and i think there are a lot
53:21of studies of this type
53:23in political science among political
53:24economies
53:26very few in economics and i i i i hope
53:29there will be there is more
53:31you know because i think we passed the
53:33time where
53:35um you know basically we think about the
53:38government
53:39as benevolence and doing what is right
53:42you know that's where generation one
53:45barrows
53:46generation but i would not like me to
53:48say that but anyway so
53:52you have to start from somewhere anyway
53:54but now you know when we look at
53:56real government with incentives
54:01to get in power discretion
54:05over public investments
54:08and hiring to retain
54:11power then we observe those kind of
54:14distortion
54:15and we need to think creatively about
54:18how you can
54:18limit this you know because
54:22um you know it's more i mean the hottest
54:25area i think in this
54:26kind of research is basically uh you
54:28know to to to find
54:30creative way you can reduce political
54:34distortion
54:35because it's natural it's structural
54:38you know there is no way somebody will
54:41be in power
54:42and not have an incentive to use his
54:45discretionary power
54:46over public spending to do this they
54:49will do it
54:50so the question like uh you know my son
54:53and chewovski will ask will it is it's
54:56going how
54:57do you design the rules and institutions
55:00in government
55:01to make this less likely to happen you
55:03know i think this is a very very
55:04truthful area
55:05that i think uh something that might be
55:07anyway so i agree with you
55:09okay so now let's talk about contract
55:11enforcement
55:13so so this is it's a it's a it's it's a
55:17really really cool paper that's just
55:19getting uh
55:20qj i think it's about how
55:23weak contrast enforcement mechanism can
55:25lead to misallocation
55:27you know so um you know and they
55:31they use a very clever kind of strategy
55:34to measure
55:35weak control enforcement strategy
55:38by by using the example of
55:42congested courts as a proxy for weak
55:45contract enforcement and they study the
55:48effect of
55:49caught congestion on firm structure
55:51productivity
55:53and expenditure and then they look at
55:56contractually what would have happened
55:58if congestion were reduced
56:00if caught congestion is reduced and they
56:02use the reduced form analysis we fix the
56:05instrumental variable and you know like
56:09basically um look at um structural model
56:13and you know and focus on parameter that
56:16governance
56:17friction so they use some you know
56:20evidence to show
56:21that um average age of pending civil
56:24cases in each court
56:26um against the age
56:30you know and and define and they use um
56:33the age as instrument
56:37for um for conjecture
56:40and uh as you can see for instance um
56:44material expenditure in a total cost
56:47negatively affected by code congestion
56:52so so that's that that's good and
56:55um anyway so to summarize
56:58so production and sourcing decision
57:00appear to be systematically distorted
57:04in states with weak enforcements you
57:07know and this is uh similar to goldstein
57:09and eudrin's paper
57:11you know where they show for instance
57:12that um chiefs
57:14in ghana for instance invest more
57:19in technology because they are
57:23land less likely to be appropriated
57:27you know so so so so status in society
57:31is your protection against um
57:34you know against um uh
57:37basically exploitation you know it's
57:39similar here so
57:40it's as if the differential contract
57:43enforcement
57:44control enforcement for the chiefs
57:46control enforcement from non-chiefs
57:48you know here the difference between
57:50states that
57:51have conjectured court states that do
57:53not you know
57:55and plants instead that with more
57:57conjectures code will shift their
57:58expenditure away from intermediary input
58:00that have a
58:01greater um vehicle plan for production
58:05in the structural model they show how
58:08weak enforcement exacerbate
58:10how the problem um you know
58:15you know anyway so just
58:19take a in-depth look at that paper you
58:21know
58:22showing how we contracting enforcement
58:25will affect um uh misallocation of
58:28resources you know i think this is
58:29this is this is the kind of um micro
58:33careful micro work linking institutions
58:36to development that i think we need more
58:39of you know because
58:41institution is broad development is
58:44broad
58:45you know so when you reduce it by saying
58:48well
58:49here contract enforcement measured by
58:53congestion court development
58:56misallocation and then you link both
59:00with data from one country you know
59:04clean you know so i think this is this
59:07is the
59:08this is very very interesting okay now
59:11the last part of the last type of
59:14political distortion
59:15which is widely widely covered in the
59:17literature currently is
59:19state captured and
59:23okay the the initial
59:27interesting and important work on state
59:30capture
59:31is due to fissman uh from um
59:35you know from bu and currently at bu who
59:38and he showed
59:39how um
59:43firms that
59:46are firms that um
59:49basically are connected to
59:53suharto how they are
59:56the you know the stock prices security
59:59prices are affected by
1:00:01news about his health so and we find
1:00:05that
1:00:06one way to predict stock prices
1:00:10and detect which one is connected which
1:00:13one is not
1:00:15is by see how the stock prices respond
1:00:18to the health
1:00:19of suharto you know
1:00:23so which shows that it showed that and
1:00:26they find that the value
1:00:28okay connected firms are valued 20
1:00:32higher because their stock prices
1:00:36drop by 20 percent if there are news
1:00:40that their connection might disappear
1:00:43by because of the health of suharto okay
1:00:47so there are other works along this line
1:00:51you know showing that this is very very
1:00:53common
1:00:55and you know there are data sets uh
1:00:58that's provided in fascio on this
1:01:00you know i'm not going to go through
1:01:01that um
1:01:04now so um now the
1:01:07the the i i think i have maybe another
1:01:10maybe
1:01:1110 minutes or so um so so let me
1:01:15just um maybe go over very quickly
1:01:18um how connections come about
1:01:23um you know and what what is
1:01:26uh you know the effect um
1:01:30so i should also before i
1:01:33i think i i should say that um
1:01:36there are several you know
1:01:39similar stories on political connection
1:01:42that i i didn't cover
1:01:44like one is by quaja and a sim showing
1:01:47for instance how
1:01:48connected firms in in pakistan have
1:01:51better access to finances
1:01:53than than other uh non-connected firms
1:01:57and they were able even to to to show
1:02:01the effect that it has on growth to able
1:02:04to quantify
1:02:05how connections and distortion
1:02:08coming from political connection affect
1:02:11growth rate
1:02:12okay so okay so i mean
1:02:15one way um connections
1:02:18happen is to campaign contribution
1:02:22there is a a great study uh done by
1:02:25about two nights 2020 by showing
1:02:29how the probability of winning
1:02:32uh contract government contract for
1:02:36instance
1:02:37uh dropped drastically when
1:02:40firms sorry when there is a ban
1:02:44on campaign contribution you see the
1:02:47probability
1:02:49of connected firms in winning um
1:02:52you know government contract is much
1:02:55higher before the ban
1:02:56and disappeared after the event it's
1:02:58extremely powerful
1:03:00evidence you know that um campaign
1:03:03contribution is one of the channel
1:03:05through which for instance political
1:03:07distortion happen
1:03:08you know and um
1:03:12so and they use you know they use
1:03:14different diff
1:03:15before and after you know it's like it's
1:03:17like a policy
1:03:18before a policy after the policy and see
1:03:21you can see for instance that
1:03:23contributes contributes and ban
1:03:26has negative sign so it shows that uh
1:03:29after the ban
1:03:31you know this gap i mean
1:03:34and this kind of study actually it just
1:03:37gives you hope right
1:03:38because when you think when you see all
1:03:41the rig
1:03:42procurement contract that you see even
1:03:44in developing countries
1:03:46you can just simply assume that this is
1:03:48structural it's not going to change it's
1:03:50there forever
1:03:52but then you bring some relatively
1:03:55modest
1:03:56kind of intervention which is limiting
1:04:00and banning campaign contribution and
1:04:03you see
1:04:04the gap be in we you see the effect
1:04:08on um on on on
1:04:12on procurement process over
1:04:15even few years you know so i think i
1:04:18strongly recommend that you you read
1:04:20this paper you might find it very
1:04:21very very inspiring because i mean those
1:04:24data exist
1:04:25you know and you'll be shocked about
1:04:28what you find
1:04:29and then you know as i said earlier you
1:04:31can be a good inspiration for you in
1:04:32terms
1:04:33in terms of how you want to actually
1:04:37i mean it can be a good inspiration and
1:04:39what kind of
1:04:40policy you know you can you can propose
1:04:43you know one thing
1:04:44is to detect um distortion
1:04:47and measure it another one is to think
1:04:49about how to fix it
1:04:50and the way we can fix it is to to to
1:04:53see
1:04:54um this kind of policy the one that i'm
1:04:55describing here
1:04:57and the mechanism through which they
1:04:59happen okay
1:05:01and um another um you know
1:05:05very important paper uh is by actually
1:05:08oxygen um for instance also
1:05:11there has some kind of competition
1:05:12flavor
1:05:14showing for instance how incumbent firms
1:05:17are entrenched they are highly connected
1:05:22they are less innovative and
1:05:26you know and the cost
1:05:29for the economy of um
1:05:32those entrenchment but again as i
1:05:35mentioned earlier
1:05:36it takes the risk of political turnover
1:05:40but it also takes institutional fix
1:05:44like the one we just described like
1:05:47banning campaign contribution
1:05:49or more transparency in government to
1:05:51reduce those
1:05:52those kind of incentives okay so um
1:05:55you know very as i said there are a
1:05:57massive number of papers
1:05:59on polycon connection on outcome i
1:06:02already talked to about kwaja and myan
1:06:04talking about uh probability of bailout
1:06:06in the u.s
1:06:07loan from state banks in brazil lower
1:06:10taxation
1:06:11easy access so great set of papers
1:06:14on showing basically providing
1:06:18evidence of political distortion coming
1:06:21from political connection
1:06:22okay this is great now one one
1:06:26before i conclude let me tell you about
1:06:28uh
1:06:29uh you know basically um
1:06:33one of the channels to which public
1:06:35connection are created is through
1:06:36lobbying
1:06:37and lobbying can be good or bad you know
1:06:40but because lobbying for instance
1:06:42is a way government can get
1:06:45expert information you know like for
1:06:47instance
1:06:49you know i mean i would like to lobby
1:06:50the us government for instance
1:06:52um to inform them about
1:06:56countries in africa that's a truly
1:06:58democratic i mean no
1:07:00assume that i know the truth right so
1:07:02truly they were
1:07:03one that's less you know and you
1:07:06lobbying for instance will help
1:07:10i don't know let's say microsoft or
1:07:12others you know to
1:07:14get the government to be more
1:07:16knowledgeable
1:07:18about you know the new development
1:07:21in the new technology you know so but
1:07:23then
1:07:24lobbying can also create
1:07:27a political connection that might create
1:07:30political distortions that i describe
1:07:32and what is interesting when you want to
1:07:34study this question
1:07:36in the u.s in particular is that it's
1:07:37very transparent you know there's a lot
1:07:39of data
1:07:39on lobbying and as i said it can be
1:07:43information information transmission
1:07:46it can also be
1:07:50provide incentives um
1:07:53incentive for political connection so
1:07:56um you know if you burn you might
1:08:00lose or you might win you know so there
1:08:02are a lot of studies
1:08:03trying to disentangle um you know
1:08:06the you know what is
1:08:10what what what effect is driven by
1:08:12information sharing
1:08:13and what is the effect is driven by
1:08:14political connection
1:08:16so there are models that
1:08:19you know look at this um you know
1:08:22so consider uh lobbying
1:08:27and how it leads to misallocation
1:08:30and so on uh particularly you know
1:08:33honest and kim
1:08:34is very very informative because not
1:08:36only theoretically but also empirically
1:08:38okay so um
1:08:41now okay let me talk about political
1:08:44application very briefly before i end
1:08:46so maybe in in you know maybe in three
1:08:49four minutes
1:08:50so okay so
1:08:53um the one implication of this result is
1:08:58that
1:08:58you know you should ban political
1:09:02connection
1:09:03and lobbying but this is not
1:09:04straightforward as i mentioned earlier
1:09:07and also in a paper that i wrote with
1:09:09nathan cannon
1:09:10um using data from benin uh we show
1:09:14that if you ban
1:09:18one form of capture you can open up
1:09:22for another form of capture because for
1:09:24instance
1:09:25government can try to capture
1:09:29true bribing politicians they can also
1:09:32capture by bribing bureaucrats
1:09:35you know they can take over
1:09:39and they can do it for instance um
1:09:44with social ties or political ties
1:09:48you see what i mean so as a result
1:09:52um
1:09:53[Music]
1:09:57you know i mean maybe i should maybe say
1:09:59more about this paper
1:10:00so we we find for instance that
1:10:04places the way there is more competition
1:10:07you know i mean in a strong pecterian
1:10:08setting you know competition government
1:10:10turnover works
1:10:12in terms of promoting and reducing
1:10:16political connection okay but we find
1:10:19that places
1:10:20where you have more competition firms
1:10:24work with bureaucrats because
1:10:28they're not supposed to be too much
1:10:30turnover with bureaucrats
1:10:31okay so places where
1:10:35you have less competition firm will
1:10:38connect with politicians you see what i
1:10:41mean
1:10:41politicians and this is linked to the
1:10:43question that was asked earlier
1:10:44you know like if there is turnover in
1:10:46government
1:10:48all right work with bureaucrats and
1:10:50agencies
1:10:51that are stable that are not changing
1:10:54you know to get your deal done if
1:10:58you are there is not competition not
1:11:01enough competition
1:11:02then work with politicians instead do
1:11:05loving politicians
1:11:06so we have to be aware that
1:11:09um we need a more comprehensive approach
1:11:11to reduce political connection
1:11:13other than just banning lobbying
1:11:15politicians because they
1:11:16actually capture bureaucrat as opposed
1:11:18to politicians
1:11:19okay so um
1:11:23anyway now let me you know let me finish
1:11:26by
1:11:27talking about several policies one is
1:11:30transparency
1:11:32so we should be talking more about
1:11:33transparency in governments not
1:11:35necessarily political turnover
1:11:37you know because and there are a lot of
1:11:39studies and audit
1:11:40government contracts and you know work
1:11:42in brazil that
1:11:43make that point we also want
1:11:46civic engagement you know for instance
1:11:48it's important that
1:11:51citizens demand transparency you know
1:11:54it's not just
1:11:55it's not enough to get your guy
1:11:58elected you know because
1:12:01turnover is not enough
1:12:04so transparent in government is the key
1:12:07you know okay but then
1:12:08transparency today is easier than before
1:12:11because of technology
1:12:12you know digitizing uh or digitizing
1:12:15yeah digitizing governance
1:12:17you know like make um
1:12:20that collections and policy
1:12:24implementation
1:12:25more transparent with uh digital id
1:12:28cards and
1:12:29whatever it's something that can show a
1:12:32lot of promise
1:12:33um you know i think we already talked
1:12:36about burning
1:12:37regulated regulatory reform and so on
1:12:40okay
1:12:41before i finish i also need to maybe um
1:12:44tell you about corny capitalism
1:12:47and so on um particularly in many
1:12:51north african and asian countries for
1:12:53instance where
1:12:54you have a state-run economy led to a
1:12:58transition
1:12:59to a capitalist system that political
1:13:03and military elites
1:13:04run the firms run the economy you know
1:13:07as a result in that case
1:13:10political distortion are structural you
1:13:13know they are not something that
1:13:14happened
1:13:15you know before and after election is
1:13:18built in
1:13:18it's built in you know and that's the
1:13:20case in indonesia
1:13:22syria and egypt and so on so it's
1:13:24something that i thought
1:13:25it's important to distinguish okay so
1:13:28okay let me
1:13:29let me talk about the future directions
1:13:31and again since
1:13:32as you can see there is still a lot of
1:13:35room
1:13:36for major improvement in in the
1:13:38literature
1:13:40is the need to develop a unified
1:13:43and yet flexible endogenous growth model
1:13:46with genetic
1:13:47generic institutions that's quantifiable
1:13:50with new
1:13:52with new data that's available you see
1:13:54what i mean so
1:13:55you see earlier i have three benchmarks
1:13:58one
1:13:59two three you know one barrel
1:14:02two uh three uh
1:14:05aguil and whatever so there might be a
1:14:08possibility to actually
1:14:09do something that cover those three
1:14:14and each of them would be like a special
1:14:16case
1:14:17you know i think that theoretical
1:14:19exercise would be very very important
1:14:20you know that's what i mean by generate
1:14:22genetic institution
1:14:24and also flexible okay so
1:14:27and that will incorporate future public
1:14:29finance control enforcement democratic
1:14:30governance
1:14:31that's our story separately now but it
1:14:33could be integrated in one model
1:14:36and um so but you know
1:14:39what something important to really
1:14:41notice that me too
1:14:43we need to be able to characterize the
1:14:45socially optimal location of public
1:14:46investment more clearly
1:14:47you know because how would you know
1:14:49distortion if you don't know what
1:14:51distortion is from you know but you need
1:14:55a model to be able to actually
1:14:57uh you know expl be very explicit about
1:15:00what it is and you need them
1:15:01anyway so that's what the model will do
1:15:03and so
1:15:04so so far we talked about political uh
1:15:07political connection as something
1:15:09that's very bad for growth and this runs
1:15:12through
1:15:13misallocation of um of resources capital
1:15:17and labor
1:15:18but it could be much deeper than that
1:15:20because for instance political
1:15:22connection can lead to misallocation of
1:15:24talent through aspiration failure i mean
1:15:26in a place where you know that
1:15:28talent doesn't matter you know
1:15:31like you are in environments where
1:15:34competency and talent doesn't matter
1:15:35in terms of working for governments or
1:15:38being successful
1:15:39you know this might dampen aspiration
1:15:43and further damage the prospect for
1:15:46growth
1:15:47so political connection is something
1:15:48which is
1:15:50pernicious deep and can be multi multi
1:15:53multi multi-dimensional
1:15:55yeah anyway so so uh that that that
1:15:58that's it
1:15:58and um you know i think we have um uh
1:16:02oh 15 minutes four question perfect okay
1:16:05thank you very much
1:16:08thank you leonard um that's perfect uh
1:16:10there's certainly some questions in the
1:16:12q a but i wanted to give
1:16:13um the people in the front row a chance
1:16:16uh
1:16:16to ask questions first abhishek i think
1:16:19you had a question
1:16:20yeah thank you professor uh
1:16:23yeah i i would take a cue from the
1:16:25burgess at all paper
1:16:27and this is related to that and
1:16:29obviously they compare
1:16:30in the distortions in public investment
1:16:33with respect to autocracy versus
1:16:35democracy
1:16:36yeah is it fair to ask is it fair to ask
1:16:40if
1:16:40within periods of democracy there exists
1:16:44some distortions
1:16:45distortions based on the relative
1:16:47strength of an ethnic or religious group
1:16:50yeah i mean we are not comparing
1:16:52opposite autopsy i'm just comparing
1:16:54within demon democracy
1:16:55yeah if it is fair to ask i mean direct
1:16:58me to some say
1:16:59empirical evidence for something like
1:17:02this
1:17:03does empirical evidence exist at all for
1:17:05something like this
1:17:07yeah i mean you know i mean okay let me
1:17:08put this way um
1:17:10uh excuse me professor sorry in that
1:17:13line too i think i just want to hear the
1:17:15question
1:17:16are there specific studies that have
1:17:17also looked at the
1:17:19level of multi ethnicity of many of the
1:17:21developing countries with respect to
1:17:23to to what we're talking about basically
1:17:27because the yeah
1:17:30in a particular country you have um uh
1:17:35uh political appointments are
1:17:36distributed aligned
1:17:39along certain um you know
1:17:42the multiplicity of uh languages ethnic
1:17:46i mean around the nations are there
1:17:48specific papers that have also
1:17:49controlled for some of these
1:17:51yeah to the practice of them yeah so so
1:17:53yeah i think okay
1:17:54this is sort of a big and broad question
1:17:57that it might be
1:17:58great to to to to to to cover
1:18:01like during the three hours like
1:18:03discussion session that we are going to
1:18:05have let me say
1:18:06two things very quickly so for the first
1:18:08question yeah i agree you know i mean
1:18:09this is very stylized
1:18:12um you know and this is one case you
1:18:14know kenya
1:18:15you know so um
1:18:19and the idea is that democracy is good
1:18:20because it
1:18:22leads to need to build a multi-ethnic
1:18:25coalition to win
1:18:27and as usual you have to deliver
1:18:28something for the other guys you know
1:18:30but then um when
1:18:34you even within democracy for instance
1:18:37i mean there are some like ethnic
1:18:39capital
1:18:40that might be explaining the discussion
1:18:43that we see
1:18:45there might be some the
1:18:49incentives some electoral incentives
1:18:53that go beyond um
1:18:56simply ethnic
1:19:00that might be explaining what's
1:19:01happening you see what i mean so
1:19:03okay just see these papers as a way to
1:19:06show that
1:19:07distortion do exist you know
1:19:10and where they come from uh have they
1:19:13been
1:19:14exaggerated and so on you know
1:19:17did something that need to be studied
1:19:19and again
1:19:20it come to the question i asked earlier
1:19:22which is that i mean okay
1:19:24that is the okay what is and actually
1:19:26very close to you please
1:19:28this is in line with your question so
1:19:31what is the baseline what are we
1:19:33comparing
1:19:34this distortion too explicitly
1:19:37you see what i mean wha where okay is
1:19:40the deviation from what
1:19:42exactly do we have a model of
1:19:45how the public the public
1:19:49investment should be made take into
1:19:51consideration not just
1:19:54the market access but many many other
1:19:56things
1:19:57the ethnic capital whatever it is you
1:19:59know
1:20:00um the state of the existing i don't
1:20:03know so
1:20:04i agree with you that more more need to
1:20:06be done
1:20:07to capture very very very
1:20:10you know in depth uh what this this is
1:20:14um now about the about the
1:20:18ethnic division tragedy literature
1:20:22um i think it's a literature that was
1:20:25extremely
1:20:25popular in the 80s and 90s
1:20:29but i think more and more
1:20:33okay more and more we are moving towards
1:20:39uh political connections more than
1:20:42ethnic collection connection per se
1:20:44you know like um
1:20:47you know like the public good
1:20:52a provision is not just driven by the
1:20:54fact that people cannot agree
1:20:56because they speak very different
1:20:58languages you know it's because
1:21:00canadians want to win election and they
1:21:02know what to do to win elections
1:21:04they will know where to invest you see
1:21:06what i mean and sometimes they invest
1:21:08in places where you have one dominating
1:21:12group in other places where
1:21:14um you know they they go to a place
1:21:17where
1:21:18the you know the votes are you know
1:21:22so i i think there is less
1:21:26focus on ethnic diversity
1:21:29today as a problem
1:21:32than patronage misallocation
1:21:36or distortion in public investment
1:21:39driven
1:21:40by political incentives
1:21:43you know and ethnicity is just part of
1:21:45the story but it's not the whole story
1:21:47that's for sure you know and um
1:21:50i mean and also you know i mean the
1:21:51story of ethnicity is
1:21:53you know i mean it's so predictable
1:21:55right i mean live you know i mean
1:21:56if you are we and i co-ethnics i'm going
1:21:58to do everything for you
1:22:00you know it's not that simple
1:22:03you know so so i think yeah it's not
1:22:06that simple
1:22:06and you know it's so mechanical you know
1:22:09so yeah
1:22:10so i think i'm not saying that it's
1:22:12totally relevant today
1:22:14but it's just to be it should be part of
1:22:16the how do we see
1:22:19political distortion you know and the
1:22:21hope
1:22:23the hope is that you know
1:22:27stories that indicate that by
1:22:30burning campaign contribution by making
1:22:33government more transparent
1:22:35you reduce political discussion you know
1:22:38and the evidence to support that
1:22:39so something can be done your ethnic
1:22:42division
1:22:43ethnic diversity hasn't changed and yet
1:22:46there are a couple of things you can do
1:22:48to reduce political distortion
1:22:50you know so and those and those new
1:22:52people that i just indicate
1:22:53i think play show that you know anyway
1:22:56so
1:22:58a couple questions from the q a um
1:23:02one is um corey vernon
1:23:06do we usually think of patronage and
1:23:08connections slash state capture
1:23:11as distinct or overlapping concepts is
1:23:14there a line that separates them
1:23:17that's the first question and then
1:23:19there's a vitantolio
1:23:21vitantonio mariella
1:23:26says very nice lecture he says could you
1:23:27please say something on the various ways
1:23:29to measure political power
1:23:31both du jour and de facto apart from
1:23:34information on politically connected
1:23:36firms
1:23:36is there a standard way for example to
1:23:38proxy for interests of relevant groups
1:23:41high-tech industries or social classes
1:23:43for example
1:23:45yes so i mean those are fantastic
1:23:47questions so
1:23:49okay so oh let me put this way
1:23:53so patronage and
1:23:56state capture okay state capture
1:24:00is firms and politicians okay
1:24:05all right patronage is
1:24:09politicians and bureaucrats
1:24:13okay basically the story is that well
1:24:17you know like in the u.s this guy
1:24:21campaigned for me in arizona i win the
1:24:24election
1:24:26uh he or she is going to become the
1:24:29ambassador
1:24:30in china you know so and there is
1:24:34a lot of papers that show that this
1:24:36patronage appointment do exist
1:24:38in in the us okay so at the beginning
1:24:41those are two separate kind of concept
1:24:43you know poly connection
1:24:45like basically it's about inside
1:24:49inside connection between a firm and
1:24:51politicians and patronage
1:24:52is when you win the election would you
1:24:54put in power to
1:24:56to do your bidding you know okay but now
1:24:58they can be connected you know because
1:25:00for instance
1:25:01what if the patron who the guy who is
1:25:05appointed
1:25:06is highly recommended by a firm
1:25:10you see what i mean like and and this
1:25:14is called revolving door
1:25:18you see what i mean so yes
1:25:21if you patronage and state capture can
1:25:25combine
1:25:27to be revolving door you know where
1:25:30somebody recommended by the firm is
1:25:33appointed
1:25:34in a government agency
1:25:38to work for the firm you know like for
1:25:39instance there is a story
1:25:41about bullying and this and the the
1:25:46uh the department of defense you know
1:25:49where
1:25:50where um you know some people say that
1:25:53when the ethiopian airlines had an
1:25:55accident
1:25:56the government didn't reply very very
1:25:58quickly because
1:26:00boeing is highly connected to
1:26:03the transportation agency
1:26:07you know so this thing happened
1:26:09everywhere you know so
1:26:10yeah yes um um
1:26:14they are separate to begin but then
1:26:16there are ways
1:26:17they can combine that that makes it even
1:26:20more more pernicious and you know so
1:26:24and then about um the power
1:26:28measuring power de facto and uh you know
1:26:31it's it's a very
1:26:32it's a very kind of um promising
1:26:36literature
1:26:37um you know so far it's uh de facto and
1:26:41it's very historical which is great it's
1:26:44also very macro
1:26:46so yeah i think in the future i will see
1:26:49like development in that area as well
1:26:51you know where for instance
1:26:54um you know the
1:26:58influence is measured
1:27:02economically you know but it could also
1:27:05be measured politically
1:27:07you know um and uh
1:27:10and sometimes influence can be measured
1:27:14formally you know in a sense like you
1:27:16know you get appointed
1:27:19um so and so it could also be
1:27:23and measured informally you know like
1:27:26for instance after all
1:27:27density opinion in china was not the
1:27:29president was not
1:27:31even the prime minister and yet he was
1:27:34the most powerful
1:27:35politician uh post mao in in china
1:27:39i mean like you know everyone else is
1:27:42just
1:27:42a figurehead you know so so what i'm
1:27:45saying
1:27:45yeah it's it's important to actually uh
1:27:48document preferences of individuals
1:27:52and see whose preferences matter the
1:27:57most
1:27:58and when and be able to distinguish
1:28:01uh to be yeah those clearly it's
1:28:03something that can be done and it just
1:28:05confirmed
1:28:06that there's still so much so much so
1:28:08much to be done
1:28:09in that area and again what is
1:28:11interesting is that
1:28:14the literature is becoming increasingly
1:28:15micro
1:28:17and there are data available you know
1:28:19historical archives and all this kind of
1:28:20stuff
1:28:21and there are a lot of contributions
1:28:23that can be made you know you take a
1:28:24simple glue and robinson's work
1:28:26or other work for instance by you know
1:28:28my other
1:28:30kind of major
1:28:34mostly macro major figure in particular
1:28:36development very macro but you take it
1:28:38with data
1:28:39take it too far you take into specific
1:28:41governments you know
1:28:42um and you apply those concepts to real
1:28:47you know kind of government and agencies
1:28:50in specific uh you know area
1:28:53you know or error you know like the work
1:28:55that um
1:28:56uh goku has done for british empire
1:29:00you know so anyway so i'm not sure if
1:29:03it's answering a question but
1:29:04you know we can we can we can follow up
1:29:06on this you know during
1:29:07um the office hours or during the
1:29:10discussion session next week
1:29:13let me give you three last questions
1:29:15which we're really out of time so you
1:29:16can decide how to
1:29:18whether and how to answer the three but
1:29:19i just want to
1:29:21give you them um the first is uh
1:29:24ismail sinusi uh
1:29:27asking about the optimal level of
1:29:29government involvement
1:29:30yeah he said one wondering if government
1:29:33input could lead to higher or lower
1:29:34transaction costs and whether
1:29:36less of such input is preferred to more
1:29:39then that's the first one the second is
1:29:43just have a first name emanuel
1:29:46well it's a specific question the
1:29:48government of uganda
1:29:50promotes oil palm production in uganda
1:29:52and uses ppp approach
1:29:55contracts a private player to offer
1:29:57inputs subsidizes and then farmers will
1:29:59help this private company
1:30:01we think that such strategy will create
1:30:02distortions that farmers might end up
1:30:04losing
1:30:05how can the effectiveness of this ppp
1:30:08contract and subsidies be studied
1:30:11and then the third question is swati
1:30:13jain
1:30:14i want to understand why the assumption
1:30:16of rational decision maker
1:30:18is not applicable for governments as
1:30:19economic agents
1:30:21does be behaving rationally increase
1:30:24their payoff
1:30:27great so those are all great questions
1:30:29so um
1:30:31you know again i will be so one thing i
1:30:34would suggest
1:30:34is i'm sure this is done already i copy
1:30:37those questions
1:30:38and send it so that i can think about it
1:30:41and answer them
1:30:42even and individually so yeah i mean
1:30:46okay the macro perspective on
1:30:48governments
1:30:49was like well you know what like
1:30:51government is god
1:30:53and then information friction blah blah
1:30:55blah leads to government to do things
1:30:57not properly
1:30:58okay i think we passed that a little bit
1:31:00so
1:31:01now it's about recognizing
1:31:05what resources of those destruction and
1:31:07think of institutions that will help
1:31:08reduce them
1:31:09you know so so i think
1:31:13i think we should move okay we should
1:31:15recognize that there is no
1:31:18perfect government to begin with we can
1:31:20have a benchmark you have a starting
1:31:22point but probably going doesn't exist
1:31:24every government is the deviation what
1:31:26is ought to be now we need to think
1:31:28about institutions to help solve
1:31:30the problem that they create that's
1:31:32becoming a mechanical design problem
1:31:34roger meyerson is is big on
1:31:37on on this and then you know like
1:31:40example of ppp
1:31:42you know like how do you minimize
1:31:45political distortion especially state
1:31:47capture by designing
1:31:49private private public partnerships
1:31:51properly
1:31:52what do you do when you have natural
1:31:55resources
1:31:56blah blah blah you know i mean so all
1:31:59those are problems that's
1:32:00that were raised and i think the issue
1:32:04is always to measure distortion that
1:32:07come
1:32:07from this environment to the best you
1:32:10can
1:32:10and then think creatively about how to
1:32:13solve it but again
1:32:15something which is clear is that by
1:32:17having more
1:32:18transparency by
1:32:22um by having more transparency um
1:32:26you know you will reduce political
1:32:29distortion
1:32:30but again you need to have a benchmark
1:32:33you know
1:32:33of what
1:32:35[Music]
1:32:36how do you actually measure distortion
1:32:38from one specific
1:32:40i mean from from what is the optimal
1:32:43government
1:32:44action that you are considering yeah
1:32:46thank you very much
1:32:47leonard um why don't the front row uh
1:32:49unmute
1:32:50and uh we'll give uh professor watchcon
1:32:52a round of applause applause
1:32:54thank you thank you thank you thank you
1:32:57that was such a pleasure so
1:32:59i'll see hopefully i see many of you
1:33:01next week yeah
1:33:02and tomorrow we have uh peak clean out
1:33:05at the same time
1:33:06um tomorrow um so
1:33:10uh we look forward to seeing uh many of
1:33:12you there as well
1:33:13okay thank you bye-bye thank you bye-bye
1:33:16bye-bye
1:33:16i should thank mandy chan as well
1:33:20i apologize kirstie and everyone else at
1:33:23cpr thank you so much for your time as
1:33:25well
1:33:25thank you

About this transcript

This page contains the full transcript of Supplemental lecture: Political Institutions and development by CEPR & VideoVox Economics, generated from the public captions YouTube serves with the video. The transcript has 11,957 words across 2,306 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.